Much of the world may currently be concerned about how to limit the impact (lack of privacy, copyright issues, job losses, world domination, etc.) of artificial intelligence. However, that doesn’t mean there isn’t huge potential for AI to improve the quality of life on Earth.

One of those applications is healthcare. With the ability to process large data sets, the deployment of AI could lead to significant advances in predictive diagnostics, including early cancer detection. While more research is needed, one of the latest studies in the field shows promising results for AI-assisted lung cancer diagnosis.

Doctors and researchers from the Royal Marsden NHS Foundation Trust, the Institute of Cancer Research and Imperial College London have developed an AI algorithm that they say can diagnose cancerous tumors more efficiently than current methods.

In the study called OCTAPUS-AI, researchers used imaging and clinical data from more than 900 patients from the UK and the Netherlands after curative radiotherapy to develop and test ML algorithms to see how accurately the models could predict recurrence.

In particular, the study looked at whether AI could help identify the risk of cancer recurrence in patients with non-small cell lung cancer (NSCLC). Researchers used CT scans to develop an AI algorithm using radiomics. This is a quantitative approach that extracts new data and predictive biomarkers from medical imaging.

Research algorithm superior to current technology

NSCLC patients make up 85% of lung cancer cases. While the disease is often treatable if detected early, the cancer recurs in more than a third of patients. The study found that using the algorithm, clinicians may eventually be able to identify recurrence earlier in high-risk patients.

The scientists used a measure called area under the curve (AUC) to see how efficient the model was at detecting cancer. A perfect 100% accuracy score would be a 1, while a pure guessing 50-50 model would get 0.5. In the study, the AI ​​algorithm built by the researchers scored 0.87. This compares to the 0.67 score of current technology.

“We then want to explore more advanced machine learning techniques, such as deep learning, to see if we can get even better results,” said Dr Sumeet Hindocha, Clinical Oncology Specialist Registrar at The Royal Marsden NHS Foundation Trust, and Clinical Research Fellow at Imperial College London, said. “We then want to test this model on newly diagnosed NSCLC patients and monitor them to see if the model can accurately predict their risk of recurrence.”

Support for practitioners – and patients

Rather than believing it will replace doctors, most now see AI in health technology as a tool that will help doctors provide the best possible care – including improved bedside manners. Despite investors gradually becoming more risk averse over the past year, the healthcare AI sector still is expected to grow from nearly $14 billion in 2023 to $103 billion in 2028.

The UK is teeming with AI health tech startups. Many are focused on drug development, genomic analysis or more consumer-oriented telehealth symptom monitoring and wearables. However, some plan to improve disease detection and diagnosis. These include the likes of Mendelianwhich has just received nearly £1.5 million to roll out its AI-based rare disease diagnosis solution as part of the government’s investment in AI technology across the NHS.

The rest of Europe also has a large number of diagnostic AI startups. Among them is Radiomics from Liège. The company focuses on the detection and phenotypic quantification of solid tumors based on standard imaging. In Norway, DoMore diagnostics uses AI and deep learning to increase the prognostic and predictive value of cancer tissue biopsies. The company’s founders also say it can help choose a therapy to avoid over- and under-treatment.

Meanwhile, a few percentage points of more accurate diagnosis, vital as they are for the affected individual, may not be the only positive impact AI could have on our healthcare systems.

According to Eric Topol, the author of Deep Medicine: how artificial intelligence can make healthcare human again“The greatest opportunity AI offers isn’t reducing errors or workload, or even curing cancer: it’s the chance to restore the precious and time-honored connection and trust—the human touch—between patients and physicians.”